Chainalysis said its investigation into an “agentic” cyber attack revealed names, usernames, password hashes and live tokens for all users — a discovery that prompted the firm to take key tools offline and lock impacted accounts while it rebuilds access controls.
Chainalysis response: services taken offline, accounts locked
In a public update, Chainalysis said it "removed access to Reactor and Data Solutions" while probing the intrusion, and that it has "no evidence that any investigations data has been accessed or Chainalysis attribution data has been manipulated." The company also implemented additional proactive monitoring and took the Know Your Transaction (KYT) user interface and API offline, alongside other address screening products and the login page.
Chainalysis warned customers that for those who "just received an email from us instructing you to reset your Chainalysis password, resetting this password won't be possible right now," and urged users to reset reused passwords on other systems. The firm asked affected customers to contact support@chainalysis.com or their Chainalysis account team with trusted IP addresses to be added to an allowlist so they can reset passwords and rotate API keys.
"We are working around the clock to harden our systems and bring Reactor and Data Solutions back online," Chainalysis said, adding that additional protections will be required to access accounts, including mandatory password resets and API-key rotations.
Scope of the exposure: credentials, tokens, and prior findings
Chainalysis disclosed that its probe exposed names, usernames, password hashes and live tokens for all users. The revelation follows earlier admissions that an "on-path attack" had previously exposed a subset of customer passwords; the company said that attack vector has since been addressed.
Out of an abundance of caution, Chainalysis has locked impacted accounts. The company said it continued to have "no evidence that any investigations data has been accessed," while acknowledging that account recovery and credential rotation will be required before services are fully restored.
This site is the portfolio.
OSINTSights runs on Cloudflare Workers, D1, R2, and Vectorize, with an AI pipeline on Hetzner ARM. Nubivance designed, built, and operates it. We do the same for clients.
See what we buildAgentic AI tools and ARTEX: the broader campaign Microsoft described
Chainalysis said the attack was similar to "many other tech platforms in recent weeks." Microsoft has separately described an agentic AI offensive framework that scans public GitHub repositories for exposed secrets — credentials, application credentials and AI inference-related API keys — and uses agents to analyze output and conduct follow-on attacks. Microsoft warned that such tooling "cuts down on the need for the threat actor to have hands-on-keyboard interactions throughout the attack chain."
Microsoft reported at least one instance in which an AI agent progressed from scanning GitHub for secrets to exfiltrating data from Azure Key Vault in just five minutes, and assessed related activity as ongoing from August 2026. The company identified sign-in events originating from two IP addresses, 199.19.77[.]84 and 199.19.77[.]18, and said, "The agent successfully conducted service principal sign-ins on dozens of associated applications across multiple tenants."
Microsoft detailed the agent’s targets — operations that return credentials such as storage account keys, App Service settings, connection strings and publishing credentials, AI and ML connections and datastores, and Azure Kubernetes Service (AKS) user credentials — and said compromised applications subsequently accessed SharePoint and OneDrive, including file downloads and mailbox-item access. The agents were also observed attempting to steal Azure resources, including compromised Azure AI services.
LLMjacking and resale: Google, Wiz and Zeabur on monetization
Google reported a growing criminal business model that steals cloud access to commercial large language models (LLMs) and resells it. Google described intermediaries called LLM Mixers that advertise access to compromised enterprise LLM accounts, proxy requests across stolen accounts and obscure the origin of access.
The Google-owned subsidiary said it identified at least two resale operations using stolen AWS Bedrock keys — llm-routing[.]com (momo) and xingjiabiapi[.]org — and noted an IP address linked to llm-routing[.]com, 23.106.46[.]239, hosted an Artex landing page in mid-October 2026 and has been flagged by Zeabur as an indicator of compromise. Zeabur previously disclosed a breach that exfiltrated over 3.2 million environment variable records and data associated with 321,810 accounts.
Wiz assessed that the emerging LLMjacking ecosystem "create[s] a market that makes the theft easily monetizable, likely driving more actors to undertake operations to steal credentials," and warned that threat actors are using stolen environments "solely to abuse their finances for mass Large Language Model (LLM) execution, incurring hundreds of thousands of dollars in AI API charges." The following IP addresses have been used to test stolen keys and are suspected to be linked to LLM token abuse and resale: 23.106.46.239, 38.246.114.118, 104.243.32.50 and 31.57.201.201.
What this means for technologists, enterprises using Chainalysis, and end users
- Technologists and security teams: expect agentic tools to look for exposed secrets across code repositories and cloud resources — Microsoft reported agents moving from GitHub discovery to Key Vault exfiltration in minutes — and anticipate that compromised service principals and application credentials can be used to enumerate and access tenant resources.
- Enterprises and procurement leaders using Chainalysis or other AI-connected cloud services: follow Chainalysis' path of allowlisting trusted IPs for restoration, require rotation of API keys and password resets, and monitor for resale markets that monetize stolen LLM access such as the operations described by Google and Wiz.
- End users of Chainalysis: expect forced account locks and mandatory resets; Chainalysis has advised users to contact support@chainalysis.com and to reset passwords on other services where credentials may have been reused.
Attribution for the Chainalysis intrusion remains unclear; the company said the pattern mirrored activity seen against other technology platforms in recent weeks. The incident sits inside a larger set of agentic-AI-enabled campaigns that target credentials, cloud-native secrets, and AI service keys — and that actors are increasingly seeking to monetize by reselling access to stolen LLMs. How quickly Chainalysis restores its analytics tools and whether the broader trend of automated, agent-led theft accelerates resale-driven abuse are the immediate facts to watch.
